Relative Sensor Pose Estimation Using Motion and Angular Shift
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing pose estimation methods for sensors are unreliable, especially when sensors of different types or with insufficient common field of view, and do not fully utilize controlled angular shift in motion.
Innovation Solution
A system and method for relative pose estimation between sensors using motion and angular shift, calculating pose shifts and angular directions within a time frame, comparing expected relations, and estimating relative pose without direct feature matching, even with low common field of view, using weighted fusion and controlled angular shifts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If direct feature matching is used for pose estimation, then the method works well when sensors have sufficient common field of view, but it becomes unreliable when sensors have different types or insufficient common field of view
Solution Approach 1:
The patent introduces an intermediary approach by using motion-based pose estimation as a mediator between sensors of different types. Instead of directly matching features between heterogeneous sensors, the system uses motion data and angular shift measurements as an intermediate representation that can be universally applied across different sensor types, thereby resolving the reliability issue when sensors have insufficient common field of view
Solution Approach 2:
The patent changes the parameters used for pose estimation from direct feature matching to motion-based parameters such as angular shift and pose shift. By transforming the estimation problem into one that relies on motion dynamics rather than static feature correspondence, the system achieves better adaptability across different sensor types while maintaining reliability
2Measurement precision
If existing pose estimation methods are used, then the process is simpler, but they do not take full advantage of controlled angular shift in motion
Solution Approach 1:
The patent applies dynamics by incorporating controlled angular shift and motion-based measurements into the pose estimation process. By making the system dynamic and utilizing temporal changes in sensor orientation and position, the patent improves measurement precision while managing complexity through structured computational approaches to track motion parameters
3Reliability
If sensors maintain fixed positions, then the system is simpler to manage, but controlled angular shift is lost which is necessary for accurate pose estimation
Solution Approach 1:
The patent applies preliminary action by pre-defining controlled angular shift patterns and expected relations between sensors before actual pose estimation occurs. By establishing these control parameters in advance, the system maintains reliability through consistent motion patterns while simplifying operation through predetermined control strategies rather than real-time complex adjustments
Data Source
AI summary
A method and system for of relative pose estimation between two or more sensors is disclosed. The system and method may include receiving a time frame, determine expected change in relative pose of at least two of the sensors, within the time frame and expected relations between angular directions of at least two of the sensors, calculating, by a processor, for each sensor a pose shift, calculating for each sensor the sensor's angular directions within the time frame, comparing relations between different sensors' angular directions, at each point in time, based on the expected relations between angular directions to determine a common direction of the two or more sensors, comparing common directions of the two or more sensors between different points in time based on the expected change in relative pose, and estimating the relative pose between the two or more sensors based on the comparison.